Let's face it.....99% of DIY synth sites and circuits out there are dedicated to the old VCO-VCF-VCA type of synths. There's only so many VCO's, VCF's, VCA's, LFO's, ADSR's that one can rehash. I'm not interested in another me-too collection, although I've been know to contribute at the Electro-Music.com website and am a proud owner of a 12u rack analog modular synth and 2X 6u rack analog drum machine. I will include a few analog synth circuits just to prove I'm a member of the team.
1. Spark Fun Wav Trigger:
This has really been my baby for the past several months. Circuit-wise, it's baby-stuff. Just get a Spark Fun Wav Trigger board, a Spark Fun midi board (if you don't want to roll your own opto-isolator circuit), a microSD card and a 9V wall wart power supply and you're in business. That's the easy part. The hard part is creating individual .WAV files for each and every note on your keyboard. Using Audacity, it takes me 1.5 hours for 4 octaves of a single voice.....and the thing will basically hold 10 voices. So that's 15 hours to set up the card.....assuming you want a full complement of voices. Fortunately, the website has a link to 5 mellotron voices already ready to load.
Why do I like this thing so much? I used to gig weekly church services and an occassional wedding or misc event. My main go-to instrument was the EMu Proteus 1. The best part of it? It had 192 different voices. The worse part of it? It had 192 voices! I really only use about 8-10 of them. I always wished I could have one that only contained the voices I used. Now I have it!
Anyway, I've been sampling soundfonts, other EMu modules, records, CDs,etc. building up my library. I think I'll need 3 cards to hold all the voices I want, which is obviously less than 30 but more than 20. ;-)
2. Raspberry Pi Soundfont Player running on Wheezy Debian Linux
I've gotten this to work with script but never put the script in an auto-start up situation. I was running full tilt on this until the Wav Trigger came along. That pretty much took the wind of the this project's sails.
Now however, there is the more powerful Rasp Pi 2 which could lend itself to some more powerful soft synths. I would like to explore this with Win 10 and see how well VSTi's work with the thing. OR....maybe a sequencer with Linux. Qtractor comes to mind. The original Rasp Pi seemed to choke on it. Another possibility would be a small digital multitrack recorder. Lots of possibility, obviously.